In 2003, a flashy little fish swam into the U.S. market, becoming the newest fad — and the newest controversy — among U.S. aquarium enthusiasts.

GloFish, as they’re called, are fish that have been genetically modified by scientists into fancier, more colorful versions of themselves by using genes snatched from other organisms, such as sea anemones. These “transgenes” — genes taken from one organism and inserted into another — code for fluorescent proteins, which are what put the “glow” in “GloFish.” GloFish’s parent company, Yorktown Technologies, has carried out this process with a handful of different species, which are now commercially available.

But while GloFish make a striking addition to your average aquarium (they come in a variety of luminescent colors, including red, blue, purple and green), concerns quickly arose — as they often do in conversations about GMOs — over how safe the little guys actually are. The federal Food and Drug Administration issued a statement shortly after their release on the market, claiming, “There is no evidence that these genetically engineered zebra danio fish pose any more threat to the environment than their unmodified counterparts which have long been widely sold in the United States.” Today, GloFish remain the only transgenic animals approved for sale to the public by the FDA.

Even so, the question has lingered until now: If they were ever to get loose in the wild, what kind of effect would they have on the natural environment?

Now, a new study published in the journal Evolution may finally put GloFish enthusiasts at ease. It turns out that at least one species — the zebrafish — is unlikely to do any damage in the wild thanks to its abysmal breeding success.

Researchers from Purdue University in Indiana were interested in how the glowing zebrafish’s transgene would affect sexual selection in a population — that is, the forces that control which individuals are able to secure mates and pass on their genes. If the transgenic zebrafish had greater reproductive success than wild fish, they might have the potential to take over whole populations. On the other hand, if they had lower reproductive success, they might get weeded out entirely.

“The idea was that … if the transgene will be whittled out of the population, then we don’t have to worry about the risks because it will disappear, and the long-term effects will be negligible,” says William Muir, a professor of animal sciences at Purdue and the study’s senior author.

There are two major forces that can influence sexual selection: mate choice and mate competition. Mate choice is usually left up to the females, who get to choose which males they prefer to mate with based on certain traits. Peacocks make an obvious example of mate choice: Females tend to prefer males with bigger, showier tails. Mate competition, on the other hand, usually occurs between males and is a form of competition in which the winner gets the mate. When deer lock antlers over a female, for instance, this is a form of mate competition.

Past studies have shown that female zebrafish prefer genetically modified (GM) glowing red males over brown “wild-type” males. But, in a confusing twist, other studies have also shown that genetically modified zebrafish tend to have less success passing on their genes, meaning something else must be stopping them from reproducing.

In order to figure it out, the Purdue researchers conducted a study in which they observed the interactions between wild-type males and GM males and recorded their reproductive success from one generation to the next.

They found that while female zebrafish prefer the showy red males, mate competition proved to be a much stronger force than mate choice. Male zebrafish compete for mates by trying to chase each other away from the females they want. As it turns out, wild-type males are more aggressive than their GM counterparts. The researchers observed a total of 7,273 chases between males, and more than three-quarters of them involved wild type males chasing red males away. When this happens, the female is basically coerced into mating with the winner of the chase, even if she would have otherwise preferred another partner.

Because of their low success rate in mate competitions, red males had much lower reproductive success and quickly died out. The researchers observed 18 different populations, recording changes from one generation to the next. In all but one population, the red GM zebrafish disappeared entirely by the 15th generation. Because wild-type zebrafish and GM zebrafish have the same likelihood of surviving into adulthood and living long enough to mate, the researchers determined that the sexual selection mechanisms they observed must be the key to the GloFish’s low reproductive success.

It’s an interesting result because scientists believe that, in most cases, mate choice and mate competition probably reinforce each other, says lead author Richard Howard, a biology professor at Purdue. But in this case, the two were shown to be at odds.

The researchers also plugged their observations into a model Muir and Howard had previously developed to help predict the risk a GMO might pose to wild populations. The model uses six life history traits, including characteristics such as adult and juvenile viability, or likelihood of survival, and male mating success, to predict what will happen to a GMO in a natural population.

When the researchers filled in the model with their zebrafish information, it predicted that the GM zebrafish would be weeded out over time: the same outcome the scientists had observed in their laboratory experiments. This result affirmed the model’s success as a predictive tool — and, in fact, the model is already used today by the federal Food and Drug Administration as a way of assessing the environmental risks of new transgenic organisms, Muir says.

As far as the zebrafish are concerned, this study shows that the GM fish would probably be quickly weeded out if any of them ever got loose in the wild. But in a broader sense, the study also provides larger insight into the mechanisms that control evolutionary outcomes.

“Genetically modified organisms have told us a lot about how genetics operate, how development operates, but they haven’t been used to really look at how evolution operates, and I think there’s vast potential there,” says Howard, the study’s lead author. In fact, Howard says, this is the first study he knows of that demonstrates the mechanisms of sexual selection at work and also shows their evolutionary outcomes.

And when it comes to GMOs, looking at these kinds of evolutionary outcomes can be an effective way of approaching conversations about their safety and their environmental risks, according to Muir. “This is such a heated and charged argument and so filled with emotion that as a scientist you say, ‘Let the data speak for itself,’ ” he says.